Say it five times fast.
Ren, our empathetic chatbot, was inspired by a desire to create a conversational agent that not only responds to user queries but also adapts its responses based on emotional cues and previous interactions. We aimed to develop a more empathetic and human-like conversational experience for those seeking mental health assistance. Using Python libraries, we built Ren with capabilities such as speech recognition, text-to-speech synthesis, and real-time emotion detection through facial expression analysis. Our custom llama2-based model tailored to mental health needs allows Ren to engage users in natural conversations, providing assistance while adjusting its responses to match the user's emotional state. Despite challenges, we're proud to have achieved milestones including successful integration of technologies for offline chatbot functionality and real-time emotion detection, and the creation of a robust architecture facilitating seamless interaction and adaptation to user emotions.
[This description was created by Ren!]
Contributions are always welcome!
LLM: Ollama
Emotion Detection: OpenCV
Storage: Local Storage
Speech: PyTTS, Speech Recognition